From: Pekka Enberg <penberg@cs.helsinki.fi>
To: Chris Wilson <chris@chris-wilson.co.uk>
Cc: Linus Torvalds <torvalds@linux-foundation.org>,
Roman Jarosz <kedgedev@gmail.com>,
A Rojas <nqn1976list@gmail.com>,
"A. Boulan" <arnaud.boulan@libertysurf.fr>,
michael@reinelt.co.at, jcnengel@googlemail.com,
rientjes@google.com, earny@net4u.de,
linux-kernel@vger.kernel.org, intel-gfx@lists.freedesktop.org,
KOSAKI Motohiro <kosaki.motohiro@jp.fujitsu.com>,
Hugh Dickins <hugh.dickins@tiscali.co.uk>,
Jesse Barnes <jbarnes@virtuousgeek.org>,
Eric Anholt <eric@anholt.net>,
stable@kernel.org
Subject: Re: [PATCH] drm/i915: Selectively enable self-reclaim
Date: Wed, 27 Jan 2010 13:20:38 +0200 [thread overview]
Message-ID: <4B602186.8050303@cs.helsinki.fi> (raw)
In-Reply-To: <1264590844-22972-1-git-send-email-chris@chris-wilson.co.uk>
Chris Wilson kirjoitti:
> Having missed the ENOMEM return via i915_gem_fault(), there are probably
> other paths that I also missed. By not enabling NORETRY by default these
> paths can run the shrinker and take memory from the system (but not from
> our own inactive lists because our shrinker can not run whilst we hold
> the struct mutex) and this may allow the system to survive a little longer
> whilst our drivers consume all available memory.
>
> References:
> OOM killer unexpectedly called with kernel 2.6.32
> http://bugzilla.kernel.org/show_bug.cgi?id=14933
>
> Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
> Cc: KOSAKI Motohiro <kosaki.motohiro@jp.fujitsu.com>
> Cc: Hugh Dickins <hugh.dickins@tiscali.co.uk>
> Cc: Jesse Barnes <jbarnes@virtuousgeek.org>
> Cc: Eric Anholt <eric@anholt.net>
> Cc: stable@kernel.org
Roman, can you give this patch a spin?
> ---
> drivers/gpu/drm/drm_gem.c | 13 -------------
> drivers/gpu/drm/i915/i915_gem.c | 22 ++++++++++------------
> 2 files changed, 10 insertions(+), 25 deletions(-)
>
> diff --git a/drivers/gpu/drm/drm_gem.c b/drivers/gpu/drm/drm_gem.c
> index e9dbb48..8bf3770 100644
> --- a/drivers/gpu/drm/drm_gem.c
> +++ b/drivers/gpu/drm/drm_gem.c
> @@ -142,19 +142,6 @@ drm_gem_object_alloc(struct drm_device *dev, size_t size)
> if (IS_ERR(obj->filp))
> goto free;
>
> - /* Basically we want to disable the OOM killer and handle ENOMEM
> - * ourselves by sacrificing pages from cached buffers.
> - * XXX shmem_file_[gs]et_gfp_mask()
> - */
> - mapping_set_gfp_mask(obj->filp->f_path.dentry->d_inode->i_mapping,
> - GFP_HIGHUSER |
> - __GFP_COLD |
> - __GFP_FS |
> - __GFP_RECLAIMABLE |
> - __GFP_NORETRY |
> - __GFP_NOWARN |
> - __GFP_NOMEMALLOC);
> -
> kref_init(&obj->refcount);
> kref_init(&obj->handlecount);
> obj->size = size;
> diff --git a/drivers/gpu/drm/i915/i915_gem.c b/drivers/gpu/drm/i915/i915_gem.c
> index 28b8f03..1ef5b54 100644
> --- a/drivers/gpu/drm/i915/i915_gem.c
> +++ b/drivers/gpu/drm/i915/i915_gem.c
> @@ -336,25 +336,25 @@ i915_gem_object_set_page_gfp_mask (struct drm_gem_object *obj, gfp_t gfp)
> static int
> i915_gem_object_get_pages_or_evict(struct drm_gem_object *obj)
> {
> + gfp_t gfp;
> int ret;
>
> + gfp = i915_gem_object_get_page_gfp_mask(obj);
> + i915_gem_object_set_page_gfp_mask(obj, gfp | __GFP_NORETRY | __GFP_NOWARN);
> ret = i915_gem_object_get_pages(obj);
> + i915_gem_object_set_page_gfp_mask (obj, gfp);
>
> /* If we've insufficient memory to map in the pages, attempt
> * to make some space by throwing out some old buffers.
> */
> if (ret == -ENOMEM) {
> struct drm_device *dev = obj->dev;
> - gfp_t gfp;
>
> ret = i915_gem_evict_something(dev, obj->size);
> if (ret)
> return ret;
>
> - gfp = i915_gem_object_get_page_gfp_mask(obj);
> - i915_gem_object_set_page_gfp_mask(obj, gfp & ~__GFP_NORETRY);
> ret = i915_gem_object_get_pages(obj);
> - i915_gem_object_set_page_gfp_mask (obj, gfp);
> }
>
> return ret;
> @@ -2580,6 +2580,7 @@ i915_gem_object_bind_to_gtt(struct drm_gem_object *obj, unsigned alignment)
> struct drm_i915_gem_object *obj_priv = obj->driver_private;
> struct drm_mm_node *free_space;
> bool retry_alloc = false;
> + gfp_t gfp;
> int ret;
>
> if (obj_priv->madv != I915_MADV_WILLNEED) {
> @@ -2623,15 +2624,12 @@ i915_gem_object_bind_to_gtt(struct drm_gem_object *obj, unsigned alignment)
> DRM_INFO("Binding object of size %zd at 0x%08x\n",
> obj->size, obj_priv->gtt_offset);
> #endif
> - if (retry_alloc) {
> - i915_gem_object_set_page_gfp_mask (obj,
> - i915_gem_object_get_page_gfp_mask (obj) & ~__GFP_NORETRY);
> - }
> + gfp = i915_gem_object_get_page_gfp_mask(obj);
> + if (! retry_alloc)
> + i915_gem_object_set_page_gfp_mask (obj, gfp | __GFP_NORETRY | __GFP_NOWARN);
> ret = i915_gem_object_get_pages(obj);
> - if (retry_alloc) {
> - i915_gem_object_set_page_gfp_mask (obj,
> - i915_gem_object_get_page_gfp_mask (obj) | __GFP_NORETRY);
> - }
> + i915_gem_object_set_page_gfp_mask (obj, gfp);
> +
> if (ret) {
> drm_mm_put_block(obj_priv->gtt_space);
> obj_priv->gtt_space = NULL;
next prev parent reply other threads:[~2010-01-27 11:20 UTC|newest]
Thread overview: 52+ messages / expand[flat|nested] mbox.gz Atom feed top
2010-01-14 13:15 OOM-Killer kills too much with 2.6.32.2 Roman Jarosz
2010-01-23 0:40 ` David Rientjes
2010-01-25 22:12 ` Roman Jarosz
2010-01-25 1:48 ` KOSAKI Motohiro
2010-01-25 20:47 ` Roman Jarosz
2010-01-26 5:19 ` KOSAKI Motohiro
2010-01-26 7:51 ` A Rojas
2010-01-26 9:06 ` Roman Jarosz
2010-01-26 11:07 ` KOSAKI Motohiro
2010-01-26 12:33 ` Chris Wilson
2010-01-26 13:03 ` KOSAKI Motohiro
2010-01-26 13:18 ` Chris Wilson
2010-01-26 13:59 ` Michael Reinelt
2010-01-26 14:07 ` Michael Reinelt
2010-01-27 0:50 ` KOSAKI Motohiro
2010-01-27 9:56 ` Pekka Enberg
2010-01-27 10:55 ` Linus Torvalds
2010-01-27 11:12 ` Pekka Enberg
2010-01-27 11:14 ` [PATCH] drm/i915: Selectively enable self-reclaim Chris Wilson
2010-01-27 11:20 ` Pekka Enberg [this message]
2010-01-27 11:30 ` Michael Reinelt
2010-01-28 3:15 ` Michael Reinelt
2010-01-28 18:21 ` Roman Jarosz
2010-01-27 11:50 ` KOSAKI Motohiro
2010-01-27 12:16 ` Linus Torvalds
2010-01-27 12:28 ` Linus Torvalds
2010-01-27 15:25 ` Chris Wilson
2010-01-27 16:09 ` Linus Torvalds
2010-01-27 17:14 ` Chris Wilson
2010-01-27 17:19 ` Linus Torvalds
2010-01-27 21:03 ` Roman Jarosz
2010-06-30 6:54 ` [Intel-gfx] " Dave Airlie
2010-06-30 7:05 ` Chris Wilson
2010-06-30 23:07 ` Linus Torvalds
2010-07-01 1:24 ` Linus Torvalds
2010-07-01 1:55 ` KOSAKI Motohiro
2010-07-01 10:15 ` Dave Airlie
2010-07-01 11:19 ` Chris Wilson
2010-07-01 22:34 ` M. Vefa Bicakci
2010-07-01 23:59 ` Linus Torvalds
2010-07-02 0:06 ` Dave Airlie
2010-07-02 0:49 ` Dave Airlie
2010-07-02 1:28 ` Linus Torvalds
2010-07-17 18:58 ` M. Vefa Bicakci
2010-07-17 19:15 ` Linus Torvalds
2010-07-18 14:27 ` M. Vefa Bicakci
2010-07-18 16:59 ` Linus Torvalds
2010-01-28 6:37 ` Willy Tarreau
2010-01-26 13:41 ` OOM-Killer kills too much with 2.6.32.2 Roman Jarosz
2010-01-27 0:14 ` KOSAKI Motohiro
2010-01-27 9:53 ` Roman Jarosz
2010-01-26 13:57 ` Pekka Enberg
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=4B602186.8050303@cs.helsinki.fi \
--to=penberg@cs.helsinki.fi \
--cc=arnaud.boulan@libertysurf.fr \
--cc=chris@chris-wilson.co.uk \
--cc=earny@net4u.de \
--cc=eric@anholt.net \
--cc=hugh.dickins@tiscali.co.uk \
--cc=intel-gfx@lists.freedesktop.org \
--cc=jbarnes@virtuousgeek.org \
--cc=jcnengel@googlemail.com \
--cc=kedgedev@gmail.com \
--cc=kosaki.motohiro@jp.fujitsu.com \
--cc=linux-kernel@vger.kernel.org \
--cc=michael@reinelt.co.at \
--cc=nqn1976list@gmail.com \
--cc=rientjes@google.com \
--cc=stable@kernel.org \
--cc=torvalds@linux-foundation.org \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox